Analytics Help CUs Manage Portfolio Risk

By Alan Veitengruber05.14.2018

Analytics enhance how credit unions view risk, but such measurements often appear convoluted and not easily relatable.

A better way to understand risk-based analytics is to think of risk in two distinct ways: default risk and loss given default.

Default risk (credit risk) normally is measured at the loan level, and is a rating that captures the likelihood that borrowers will not repay their full loans. That means the credit union may lose principle and interest payments if the borrower becomes delinquent.

Although credit scores act as the industry standard in quantifying credit risk, you may also want to consider these questions when building your default risk model:

  • How much has the borrower’s credit score migrated since origination?
  • Is there a co-borrower on the loan?
  • Is the borrower currently delinquent on this or another loan?
  • Bankruptcy navigation indexes or debt-to-income estimators.

Loss given default, measured in dollars, represents the value at risk if a default occurs. Loss given default risk is tied directly to collateral securing the loan and represents the recoverable amount if a loan is charged-off.

This dovetails with an important distinction: a default on its own does not result in a loss.

To experience a loss there needs to be both a default and a collateral deficiency. A collateral deficiency occurs when the loan balance at default is greater than the value of the collateral.

To quantify your credit union’s loss given default, you will want to know:

  • What is the current loan-to-value (LTV) ratio?
  • If the borrower has a line of credit, what would the LTV be if the line was fully funded?
  • If the loan is a second mortgage, is there a superior loan?
  • What is the loan’s current default risk?

Intuitively, the loss given default on a high default risk loan can give you a measurement to understand how the risk in your portfolio is changing over time. This intersection is where your future losses are most likely to occur.

For members with high default risk, it may be beneficial to manage collateral risk to minimize instances where a default results in a loss. This includes setting LTV limits for new loans that minimize potential losses due to normal changes in the value of collateral.

Conversely, for loans with higher collateral risk it may be important to manage credit risk. This is particularly important in uncollateralized loans or second mortgages that have a large superior loan balance.

In these scenarios, a complete picture of the collateral position is not enough. A reliable default risk model will reveal loans where weak credit fundamentals could result in a default.

Understanding the different drivers of default risk and loss given default can give your credit union a new appreciation for loan analytics and allow you to use your data more meaningfully.

A sound default risk model will also give you a better understanding of your members and the products you provide.

Alan Veitengruber is senior analyst with Twenty Twenty Analytics, a CUNA Strategic Services alliance provider. Reprinted from CUNA News.